Incorrect depth on imported dive

[steve schiff]

Subsurface 4.9.7

Windows 10 Pro

Oceanic VT Pro

Dive 524 in my logbook shows a depth of 2179'. It should be 121'. Is there any way to fix this errant data?

Thanks

Steve

[Jason Bramwell]

What does the dive log look like? Does there appear to be a spike where you plummet into the abyss or does it look ok?

Can you possibly export that dive and attach it here? It may be possible to just edit it manually.

Jb

Sent from my iPhone

On 2 Mar 2021, at 03:02, steve schiff <idocs...@gmail.com> wrote:



--
You received this message because you are subscribed to the Google Groups "Subsurface Divelog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to subsurface-dive...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/70b3d95f-42bb-477f-9b31-ad3b33feea3fo%40googlegroups.com.

[Jef Driesen]

This is a known problem with the VT Pro. There is some occasional data
corruption happening during the download. The weirdest part is that we can't
detect it, because the packet checksums are all correct. So somehow the
corruption already happens before the checksums are calculated and the packets
are send out. But it's certainly there, because if you download the same dive(s)
twice, you can clearly see they are not always binary identical.

Depending on where exactly in the data it shows up, it can result in all kinds
of issues. Most often it's somewhere in the profile data (because that is
usually the largest amount of data), which results in some outlier value (for
example a depth spike).

Unfortunately there is not much we can do about this. The only advice I can give
is to remove the problematic dive from your subsurface logbook and try to
download it again. The problem seems to happen at random, so there is a good
change you get a good copy now.

Jef

[steve schiff]

I've had this issue in the past and have seen depth spikes on the dive profile and one of the subsurface programmers was able to assist me in editing the dive, I was hoping for this again since there was an incident in this dive that led to a DECO situation and I wanted to review the profile.

I carry 3 VT Pro computers on every dive and they all have the same profile errors in subsurface. I even tried creating a new logbook to no avail.

To the poster above that suggested I post the exported file, here it is attached to this post.

Thanks for your continued assistance with this.

- Steve

[steve schiff]

I tried to attach the exported file but I keep getting "error posting message" unless I remove the attachment.

I can message it or else please advise on how to handle this issue.

Thanks

[JB2Cool]

  Feel free to send to me personally (either just the problematic dive or the entire log book) to see if that help and i'll take a look, if that doesn't work for you then put the log file somewhere and send me a link and i can grab it. You might be able to edit it in the GUI but you might have to resort to editing the raw log file.

Jason

--
You received this message because you are subscribed to the Google Groups "Subsurface Divelog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to subsurface-dive...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/68b6e60d-34e1-457d-b91d-9ac32e434d89n%40googlegroups.com.

[Jef Driesen]

On 2021-03-02 10:09, steve schiff wrote:
> I've had this issue in the past and have seen depth spikes on the dive
> profile and one of the subsurface programmers was able to assist me in
> editing the dive, I was hoping for this again since there was an
> incident in this dive that led to a DECO situation and I wanted to
> review the profile.
>
> I carry 3 VT Pro computers on every dive and they all have the same
> profile errors in subsurface. I even tried creating a new logbook to
> no avail.

Do you see the exact same error in the data from all three dive
computers? In that case it's very unlikely to be related to data
corruption issue, because that would be more random (e.g. affect only
one of the three profiles, or errors at different places).

Jef

[steve schiff]

Yes it's the same depth error on all 3 dive computers.

To JB2Cool please advise how to send you the logfile directly, and exactly which logfile you need.

I'm not seeing how to do this.

Thanks

[Jef Driesen]

On 2/03/2021 16:41, steve schiff wrote:
> Yes it's the same depth error on all 3 dive computers.

In that case, can you send me a logfile and memory dump of all three dive
computers to have a look at the data? In the subsurface download dialog, enable
both the "Save libdivecomputer logfile" and "Save libdivecomputer dumpfile"
checkboxes.

Jef

[JB2Cool]

On the desktop application File>Export should take you to the right place, you can then choose either selected dives or all dives and save these to a file. Simply email that file over to me like you would for any other file (or put it somewhere and send the link). If you send over the log of all the dives you'll need to tell me which dive(s) i need to look at. You need to be using the desktop app though, not the mobile app.

If you can see the depth spike on the dive profile you can likely fix this yourself by selecting the dive and selecting Log>Edit dive in planner and then simply dragging the little spots so that they look better before pressing save.

What Jeff is trying to do is to find out where these errors are coming from to stop them happening again.

Jason

--

You received this message because you are subscribed to the Google Groups "Subsurface Divelog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to subsurface-dive...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/c741b62a-9352-4bb0-8740-73d24b3d9fc5n%40googlegroups.com.

[steve schiff]

Jason how do I find your email address?

[Jason Bramwell]

Jb2...@gmail.com

Sent from my iPhone

On 2 Mar 2021, at 20:40, steve schiff <idocs...@gmail.com> wrote:

Jason how do I find your email address?

To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/e4b39ae4-a626-470d-81b3-ed1b9608ffdcn%40googlegroups.com.

[steve schiff]

Much thanks to Jason for editing the logfile and fixing the issue.

[JB2Cool]

For completeness all i did was edit the log file to 'fix' the issue but i can't explain how the issue came about. Jeff might be right that this was somehow download corruption. Looking at the dive log in a text editor it clearly states that <depth max='664.159 m' mean='31.137 m' /> but i can't see any samples that go over 40m so no idea why it thinks this is the max depth.

JB

To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/177a31a5-fe61-462c-a373-6abafb98101fn%40googlegroups.com.

[Linus Torvalds]

On Wed, Mar 3, 2021 at 12:52 PM JB2Cool <jb2...@gmail.com> wrote:
>
> Looking at the dive log in a text editor it clearly states that <depth max='664.159 m' mean='31.137 m' />

664.159 m is 2179 feet.

And that is 1000 1000 0011 in binary.

I bet that the high bit is garbage, or means something else (ie it
might be a warning bit).

So it's probably _should_ be 1000 0011 (binary) feet, ie 131 ft, which
is 39.9 m.

Which sounds about right if the target depth for that dive was 40m? If
the mean depth was 31 m, a max depth of 40 m certainly doesn't sound
entirely unlikely.

Of course, it's also very possible that if it was some download error,
more than one bit was wrong (in fact, simple error correction models
like parity basically would _require_ that more than one bit was
wrong).

But it sounds more likely that the dive computer doesn't really
support a full 12 bits of depth, and the high bits mean something else
(maybe a warning bit)?

Jef?

Linus

[JB2Cool]

39.9m looks absolutely right based on the log.

--
You received this message because you are subscribed to the Google Groups "Subsurface Divelog" group.
To unsubscribe from this group and stop receiving emails from it, send an email to subsurface-dive...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/subsurface-divelog/CAHk-%3DwjFCkTJYrb3%2BtYJ2MiNwLD7ze6T3rBPFxexck%3DNLhf_%2Bw%40mail.gmail.com.

[Jef Driesen]

I haven't looked at the data yet, but I alreadhy suspected something like this.
If all three dive computers show the same error, then it's nearly impossible to
be some random bitflip. As you say more likely some bit that got set that we
haven't seen before. I'll take a look at the data tomorrow (it's already getting
late here).

Jef

[Mikael Lindstrom]

Hi,

I recently started using subsurface (great program), and I have a Aeris Elite T3, i.e. the same manufacturer and I think file format as the original poster.

I have a similar issue where there are some errant entries (either a lot shallower (even depth=0) or deeper than the data points around them) - this happens both for dives downloaded by subsurface directly from the dive computer but also when I imported Oceanic/Aeris files that I had created using Oceanic's/Aeris' standard program (Aeris Computer Interface").

* So the download errors are not unique to subsurface, it seems.

* Pretty much every single dive has these incorrect data points.

* I do see incorrect data points when I display them in Aeris Computer Interface program.

I can probably live with these data points in the graphs - the only issue is that they mess up the summary and statistics :-(

/Mike Lindstrom

diverguy